DoG's Top 100 Test Batsmen: Discussion thread

[trundler]

Give out points for numbers of balls faced

 

[Days of Grace]

Batsmen like Gilchrist should be rewarded for their counterattacking innings where they totally changed the course of the match and put the pressure back on the opposing team.

 

[Burgey]

Word out of the Indian Cricket Journalism thread is Sehwag > Bradman, so it’ll be interesting to see how this plays out in DoG’s rankings.

 

[TheJediBrah]

Give out points for numbers of balls faced

Which would simply cancel out points from Strike rate and hence make both measurements a complete waste of time

 

[trundler]

Which would simply cancel out points from Strike rate and hence make both measurements a complete waste of time

I'm sure there's a way to weigh both appropriately in conjunction with one another. I don't have it but there's enough nerds here to do the math.

 

[TheJediBrah]

I'm sure there's a way to weigh both appropriately in conjunction with one another. I don't have it but there's enough nerds here to do the math.

No, if using them as a differentiating measurement then mathematically they will directly cancel each other out. Surely that is obvious.

 

[trundler]

No, if using them as a differentiating measurement then mathematically they will directly cancel each other out. Surely that is obvious.

You're just not trying hard enough.

 

[stephen]

You can always pick a random strike rate (say 60) and reward batsmen for being as close as possible to that strike rate. That's how absurd rating based on strike rate or based on balls faced is.

Although, perhaps a reward can be given for number of Cowans, regardless of runs scored - a bit like hundreds or fifties except not quite as good.

 

[jaideep]

RPI awards batsmen who made contributions to their team’s total run tally. It is especially important in rating batsmen in ODIs.

I’ve decided to go with Average*4 RPI *2 Strike-rate *1

Average is far more important than RPI and RPI is far more important compared to strike rate.I would suggest Average *8 RPI *3 Strike rate *1 ratio.

 

[jaideep]

Hi all,

I will in the coming months present my top 100 test batsmen of all time using my standard countdown formula.

Before I do so, I want to make this countdown list more interactive, in that I want my fellow posters to decide on what measures I should use and how much weight to give each measure.

The measures I used for the top 100 test bowlers were as follows:

Wickets taken/years active/great bowling performances 10%
Career record 50%
Peak record (best 50 innings block) 20%
Non-home record 10%
Quality opposition record 10%

The career, peak, non-home and quality opposition records consisted of average, strike-rate and points-per-innings (using my individual innings performance metric).

I want the formula to be consistent for both batsmen and bowlers. I have decided to do away with the points per innings as it gives too much of an advantage to batsmen and bowlers who scored a lot of centuries or took a lot of five wicket hauls in their career. Instead, for batsmen, I only want to use average and strike-rate (with a ratio of 4:1, the same as for bowlers).

The three questions I have are:
1. Should I keep the same weights for each measure? I feel that peak, non-home, and quality opposition should have the same weight.
2. Should I drop the peak 50 innings measure and instead use a peak career record (i.e. when a batsmen achieved their best career average)?
3. Should a player get their full rating after only 50 innings (or about 30 test matches)? Or should it be 100 innings? I feel this works better for ODI players.

I look forward to your opinions. Cheers.

Having a better peak record means the batsman was less consistent/good in the non-peak matches either side of it.20% for peak record is too high.60% for career record and 10% for peak record would be better.

 

[Coronis]

Batsmen like Gilchrist should be rewarded for their counterattacking innings where they totally changed the course of the match and put the pressure back on the opposing team.

Yes, but should players also be punished for seeing off the new ball or scoring 40 (250) to save a match? Dunno if you can have one without the other.

 

[Groundking]

Why are you even bothering taking S/R into account? It can only ever be useful in the context within a match as to the quality of that individual knock, career strike rates means nothing for how good a batsman was or not.

Unless you go into every single match and evaluate how much the batsmans strike rate added of subtracted from the quality of the knock, which is a LOT of work, it doesn't tell you anything

 

[Days of Grace]

(Pedantry here but Pollock contributed more runs to his test team than Cairns. I get your point though).

Pollock
Average: 32.31
RPI: 24.24

Cairns
Average: 33.53
RPI: 31.92

Huge difference in RPI despite similar averages means that Cairns was a more valuable batsmen who contributed more to his team which is what I want to measure along with a batsmen’s pure ability (average).

 

[Malcolm]

For the purposes of this exercise, I will not use the obviously flawed batting average metric. Instead, I will use the Weighted Batting Average (WBA). In order to negate the huge disadvantage faced by top-
order batsmen when it comes to calculating the average, in the WBA:

All innings ending in dismissals are counted as 1.

All not-out innings above the average Runs per Innings (RPI) is counted as 1.

All not out innings below the average RPI are assigned proportionate innings values between 0 and 1.0.

Home and away: batsmen who travel well and those who don't | The Cricket Monthly | ESPN Cricinfo

This metric has a value between the Batting Average and the Runs per Innings. Nearer the Batting Average for those with fewer not outs and nearer the RPI for those middle-order batsmen with many not outs.

 

[Days of Grace]

You want to deduct points from someone like Sobers who batted lower?

Hutton
Average: 54.10
RPI: 48.22

Sobers
Average: 54.90
RPI: 47.69

Very similar RPI and averages despite totally different batting positions. Sobers wins here because he has the better average, the dominant measure used in my analysis.

Openers also have chances to get low unbeaten scores when their side is chasing a small total in the 4th inns.

 

[Coronis]

Fun (though unrelated) fact, Joe Root has the lowest RPI of all batsmen who average 45+.

 

[MrPrez]

Pollock
Average: 32.31
RPI: 24.24

Cairns
Average: 33.53
RPI: 31.92

Huge difference in RPI despite similar averages means that Cairns was a more valuable batsmen who contributed more to his team which is what I want to measure along with a batsmen’s pure ability (average).

Is being valuable to your team something that should be considered when judging the best Test batsmen? Surely this list intends on measure batting ability rather than value to side. It's not Ricky Ponting's fault he batted in the dominant Aus team while Andy Flower batted for Zim, for example, and it shouldn't realistically affect where we place them as all-time batsmen.

 

[ankitj]

Pollock
Average: 32.31
RPI: 24.24

Cairns
Average: 33.53
RPI: 31.92

Huge difference in RPI despite similar averages means that Cairns was a more valuable batsmen who contributed more to his team which is what I want to measure along with a batsmen’s pure ability (average).

If Pollock played for NZ, he would have an RPI similar to Cairns. For evidence, see their averages.

 

[TheJediBrah]

Huge difference in RPI despite similar averages means that Cairns was a more valuable batsmen who contributed more to his team which is what I want to measure along with a batsmen’s pure ability (average).

If you insist on believing this then by all means put in in your analysis, it is after all your analysis. But don't be surprised if most people think it's strange, because it doesn't really make any sense.

 

[stephen]

I think career strike rate can be taken into account as well as the number of Cowans a batsman gets. If the batsmen are rewarded for facing 100 balls but are also rewarded for a high strike rate them isn't that what we'd love to see in a batsman - someone who strikes quickly and occupies the crease? Consistency and aggression?